Detailed text transcripts for TV channel - CNN - 20190217:15:43:00
Than human beings. are there other areas? it feels like this could be applied to other things. >> i think diagnostics is perhaps the most obvious one. but maybe more interesting are some of the less obvious applications like the work we started to do on therapeutics. therapeutics is not in itself an obvious prediction task. it's more of a design problem. you're looking to design a therapeutic, a treatment that has a particular therapeutic effect on a person. and so how do you define that even as a machine learning problem? where do you get the data to train a computer to make those predictions...